Poor Receive From RTL-SDR R820T2 RTL2832U 1PPM TCXO

[CanesFan95]

So I'm new to SDR and just got two of these:

RTL-SDR R820T2 RTL2832U 1PPM TCXO

With a handheld scanner on a rubber duck in the living room, I can get the local NOAA weather full scale crystal clear. But both dongles receive it poorly using the supplied telescopic antenna. The dongles actually come with 2 telescopic antennas, a short little one, and a 5-foot long one.

Even with the 5 foot antenna fully extended and moving it around, the signal is a lot weaker than the scanner and the audio out of the computer speaker is snowy / hissy like a weak RF signal.

Is this normal? Do regular scanners receive better than dongles? Everyone is saying how good this dongle is and both of mine are a big dissapointment. This is using SDR-RADIO software, which I guress is also called SDRConsole?

 

[Markb]

On the home tab, on the far right is the gain control. That should be increased probably. Mine is at about 40db.
Try that for starters.

Sent from my SM-G920V using Tapatalk

 

[Markb]

The button that says "RF".

Sent from my SM-G920V using Tapatalk

 

[CanesFan95]

The button that says "RF" was set to automatic. If I put the gain all the way up, I get no real improvement. The most it goes up to is 49.6 dB.

 

[Markb]

Ok. Keep that set to automatic and try turning off AGC and adjust the gain slider manually and see if that helps.

Sent from my SM-G920V using Tapatalk

 

[CanesFan95]

The AGC is set to medium, but everything is greyed out and it won't let me turn it off or change anything. Which slider is that that I would use?

Thanks.

 

[Markb]

Well, back up a second. Do you have the mode and filter width set properly? Do you have a strong signal on the waterfall? Check those out first.

Sent from my SM-G920V using Tapatalk

 

[CanesFan95]

I have it set to FM narrow at 12 kHz. There is a signal on the waterfall, but I'm not sure if it's considered strong.

 

[Markb]

The other issue is likely the antenna.

Sent from my SM-G920V using Tapatalk

 

[Markb]

A screenshot would be helpful also.

Sent from my SM-G920V using Tapatalk

 

[CanesFan95]

Here's a screenshot.

 

[CanesFan95]

I found an SMA adapter lying around and slapped a 5/8 wave 2m mag mount antenna on the dongle. It improved the receieve a little bit and now it sounds clearer, but it's not showing full scale. I'm wondering if both antennas I got are duds, or if maybe this dongle is not as sensitive as everyone says it is.

 

[popnokick]

Actually, you had a VERY strong improvement in signal when you changed from the simple telescopic whip antenna to the 2M 5/8 wave mag mount. For the 162.550 NWS signal you went from -78 dBm to -54 dBm. That's a 24 dBm improvement, which is huge. Incidentally, the NWS WX broadcasts are FM-Wide (24 kHz) not FM-Narrow (12 kHz).

 

[br0adband]

A few questions:

- how far are you (if you happen to know) from where the actual NOAA broadcast is coming from? In my experience they're usually from an airport or close to one since those tend to operate 24/7 (at least the big busy ones are)

- what brand/model of handheld scanner are you using where you state you get a "full scale" signal on that frequency?

As for tips:

- turn off the Automatic gain, set it for something around 29.7 dB to 36.4 dB or somewhere in there, maxing it out or using the Auto setting is impractical for FM signals since it tends to cause RTL sticks to start having image/ghosting/intermod issues which is just bad all the way around.

- you appear to have two actual VFOs created (A and B) but on the display on the right side you've got all 6 potential ones (from VFOA through VFOF) showing up there in the same area so, my advice - because SDR-Radio aka SDR-Console can be a daunting app with a lot of complexity, get rid of anything you're not actively using at a given time; in other words, hide those additional VFOs if you're not actively using them and even get rid of the VFO-B as well unless you have actual need of it and actively use it

- I'm guessing you've adjusted the ppm as required even in spite of it being a TCXO version of the RTL sticks, it doesn't hurt to verify it against a known frequency like that one at 162.550 which is the same frequency we have here in Las Vegas, I can also receive NOAA at 162.400 but the signal is much weaker (but still readable at 36.4 dB gain) as noted in the screenshot

- I'm not sure why your spectrum(s) appear to be almost maxed out but that's not good IMO, it might be more helpful to adjust them so your noise floor appears at the bottom of the spectrum as required but that's a more personal setting so choose whatever works best for you

I'm getting a signal level that tends to show about -73 to -75 dBm on the SDR-Radio signal meter and it's damned "LOUD" - I have the volume on the app set to 5 because it's just booming but if you look at the actual spectrum. SDR-Radio does things a bit differently than other SDR apps I've used, so I can't say exactly what the levels are as if I tune the same thing in SDR# (using all the same hardware and the same 36.4 dB gain) it shows a peak of -15 dB and the signal is just as clear and booming in terms of audio quality.

Noise floor on these RTL sticks is typically -60 dB (I have the original R820T style, not the newer R820T2 which I believe has a slightly lower SNR around what, -65 to -70 dB from what I've read) so that's what I use as the basis meaning I adjust the spectrum so it gets that noise floor just above the bottom. This ensures that what I'm seeing is primarily signal and not a waterfall full of nothing but noise - I only care about the signals, not the noise.

Unless you're LOS with a tower broadcasting the actual NOAA signal or in a very close proximity you're going to get a signal but it might not necessary come in anything like a local broadcast will that is coming from a tower or antenna installation that's much closer.

Also realize that these are "cheap USB TV tuners" that luckily were discovered to work - not as well as a dedicated device like a scanner or communications receiver - as radio receivers. They won't match what a dedicated device can do but they will perform acceptably for most any general purpose. It looks like you're getting a signal so that's a good thing but of course these types of devices are not nearly as sensitive as a dedicated device can be.

And another thing: those antennas they include are still "bottom of the barrel" in terms of production quality, especially the cabling: it's cheap as hell and that's part of why the cost(s) for these devices are so low. You proved that to some degree already by attaching what is probably a better antenna, and the antennas that came with the RTL sticks you got don't have a decent ground working with them either.

They work, they just don't work great, but acceptably in most situations.

Here's what mine looks like presently:



And with SDR#:



I'm using an R820T stick fed into an MCX to BNC 6" pigtail which is currently fed into a "bazooka" dipole I made earlier this morning just as an experiment - it's basically 23.25" of a piece of coax (I don't even know what type it, to be honest, it was provided by Cox Cable years ago and I happened to have some excess piece of it) with the center conductor exposed and the braid rolled back over the sheath then stuffed into a piece of PVC hanging from a curtain rack in my apartment.

The antenna works better than I expected, actually, but it's "cut" for 127 MHz to help improve my civ air band monitoring just as an experiment, as noted.

Anyway, it looks like your setup is working, it just isn't going to perform nearly as well as the scanner will in most situations. Get some better cabling for the antenna and things will probably improve, and obviously try different antennas as well. I recently re-assembled the OCFD I made about a year ago and for whatever reason now it works as I suspected it should have worked the first time but didn't. That's my general purpose antenna nowadays and it's basically standing right next to the bazooka one I made this morning.

Ain't perfect but they work...

 

[prcguy]

I think Br0adband hit on some important facts that the USB dongles are very poor performers compared to a typical police scanner. Compare the noise floor of -60dBm or even -70dBm of an RTL dongle to maybe -115 to -120dBm for a police scanner. That's a huge difference in dynamic range that will result in the police scanner picking up weak or marginal signals that the RTL dongle simply cannot hear.

The RTL dongles are a miracle $10 device with a lot of free high end software available, but they are still just a toy.
prcguy

 

[SCPD]

prcguy: You also nailed it right on the head. I would not use SDR-Console as my program of choice for a RTL dongle. I would use HDSDR. Most people think that these dongle are just plug and play and where the hell are all the signals. It's not. It does have a learning curve. The RTL is sensitive but as you pointed out they do not have the dynamic range.

The best thing is to shut of anything in the rtl control that references AUTO.
I know this reply is not much help but if you want to try HDSDR i can get you setup in 3 minutes.

 

[SCPD]

BTW the rtl-sdr dongles are the best ones out right now... You did get something decent to get your feet wet with SDR's.

 

[CanesFan95]

Actually, you had a VERY strong improvement in signal when you changed from the simple telescopic whip antenna to the 2M 5/8 wave mag mount. For the 162.550 NWS signal you went from -78 dBm to -54 dBm. That's a 24 dBm improvement, which is huge. Incidentally, the NWS WX broadcasts are FM-Wide (24 kHz) not FM-Narrow (12 kHz).

I tried the other 2 FM-Wide settings with a few different filters and that made the WX audio sound a lot quieter. I've always thought it's odd how the signal strength meter has a negative number. I've never understood what a "dBm" is and why evidently a dBm of 0 would be an incredibly strong signal.

 

[CanesFan95]

The NOAA transmitter is about 25 miles away. The scanner is a PSR-800 and it shows all the signal strength bars lit up on the NOAA frequency (162.550). I haven't adjusted the ppm and barely even know what that means. I'm not how to tell which way to adjust it and how much. I've heard of SNR before but am not sure how to interpret the unit of measure "dBm" and what amount is considered good or bad.

Is it better to get a USB extension cord instead of a pig tail?

 

[br0adband]

What we're saying is "it appears to be working as it should," more or less. You've got a signal, a better one using the better antenna obviously, so as long as you can hear the weather then yes, it's working. There's no need to go crazy trying to figure it all out, some of us have been using this type of hardware for a few years now and no, you don't understand all of it in the first few mins, or hours, or even days in some situations.

dBm is basically a way to measure signal strength using a standard figure to work with which in the case of that specific one - dBm - the little "m" there means milliwatts as opposed to dBW which would be watts and far far greater signal power.

SNR = Signal to Noise Ratio and the higher it is the better meaning you have more signal than noise. Basically if you're listening to a very weak signal and your receiver has a high noise floor - as noted, these RTL sticks typically have -60 to -70 dB noise floors meaning if you attach an antenna and tune to nothing in particular, you should see a "flat line" from one side of the spectrum to another at about the -60 to -70 dB level which represents noise (and everything below it is just a literally blanket of nothing but noise. The idea is that you want a signal to get through the noise and appear on the spectrum as a peak, and the higher that peak is compared to the noise floor the better the signal will be with respect to receiving whatever is being broadcast.

The pictures you posted, especially the one with the 5/8 wave antenna attached, show that yes you're getting a decent signal: the -74 dBm signal isn't that different from what I happen to be receiving too. Obviously the 5/8 wave boosted that signal reception considerably as prcguy pointed out when it showed -54 dBm which is a fairly massive jump in signal power that the RTL stick is getting so obviously it's going to be able to provide you with clearer audio as well.

Anything you can do to increase the signal strength aka the peaks on the spectrum without boosting the noise floor at the same time is the goal of pretty much every person that is part of this hobby. It's not always the easiest thing to do with these RTL sticks considering a) they have a relatively high noise floor to begin with in comparison to dedicated hardware like scanners and so forth as already noted and b) they are not shielded at all as they could or should be - they're in plastic cases and that doesn't help much at all in terms of RFI reduction.

Wrapping your RTL stick(s) in aluminum foil can help keep the noise floor down low where you want it, but it's not an absolute requirement to do that. There are some nicely constructed aluminum cases you can get nowadays too for most RTL stick designs (some use MCX connectors on the side, some use the SMA connectors on the end, I've seen some with F connectors on the end as well so there's a lot of variety). Shielding the RTL stick even with some aluminum foil wrapped around it can't hurt (it might make it get warmer but with the TCXO on the sticks you have will keep frequency drift at bay with nothing more than 1 ppm at best).

When I suggested you verify the ppm I was simply saying you should verify you're on frequency by checking it against something like the NOAA broadcast but since you have sticks with the TCXO then it's not really going to be a problem for you. My old 1st gen R820T sticks have pretty average drift, one is about 55 ppm and the other is about 48 ppm so that means when I tune to a specific frequency unless I account for that with the software then it'll always be slightly off the frequency I want to listen to. That's not so bad with analog transmissions like typical voice calls on radio systems, but with digital signals in some instances even just 1 ppm can cause your decoding of them to fail.

There's a lot to learn with this SDR side of doing things, and I could go on for hours providing info but I'm gonna cut it short. Most of my posts tend to be long and smack of tutorials but, the info you want to know is "out there" meaning it's in a lot of threads here in this subforum at RR and also the Voice/Control Channel Decoding Software subforum as well. It's not that tough to get going but yes it can prompt a lot of questions - they've all been asked before, dozens of times so it might do you more good to spend time reading through the threads that already exist.

There's also a book or two available now about RTL-based SDR listening specifically, they're both on Amazon so if you do a search there for SDR you'll find 'em.